Philosophy
Assessment in SoCIETIE is designed to enable your learning journey - not to test you against a fixed standard.
No Right Answers…
Complex, transdisciplinary problems don’t have single “right” answers. They need multiple perspectives and can be solved in different ways. This creates a challenge: how do we assess your work fairly when there isn’t one correct solution?
Instead of traditional grading, we’ve built our learning around these principles:
- connection: Learning happens together - with peers, educators, and the real world. Your ideas matter and are valued.
- engagement: We learn best when we’re genuinely interested. We create a safe space where different perspectives are explored and knowledge is built together.
- inspiration: Learning should drive us to create meaningful work and contribute to real-world change. Quality comes from intrinsic motivation, not external pressure.
How We Learn — the principles as a slide deck
The three above are the short version. The deck walks through the full set of principles behind SoCIETIE assessment, and states the assumptions behind each one as a provocation for discussion.
This course uses a ‘valuation’ structure rather than a traditional ‘marking’ or ‘judgement’ structure.
Feedback Instead of Marks
During the semester, you won’t receive numerical marks (like 7/10) on your work. Instead, you’ll get structured feedback using quality descriptions. Here’s why:
-
Complex work needs context: A single number can’t capture the nuance of transdisciplinary thinking. Your work deserves more than a score.
-
We assess the whole, not the sum: Your learning isn’t just Assignment 1 + Assignment 2 = Final Grade. We look at how your ideas develop and connect across the semester as a whole.
-
We want you to take risks: If we immediately score and rank your work against traditional standards, it discourages exploration. We need to support you trying new approaches, even if they’re uncertain at first.
How Your Final Grade Works
The approach to your final grade depends on which version of the course you’re taking.
In AATD, your final grade will be “Course Requirements Satisfied” - meaning you’ve met the course standard. This isn’t a pass/fail at 50%. We expect quality work at a Distinction level or higher. If your initial submission falls short, we’ll give you feedback and ask you to revise. Resubmit with these notes in mind, and you’ll meet the mark. We’re here to support you reaching the standard, not to gatekeep grades.
Course Requirements Satisfied: Hurdle tasks and optional tasks use the following feedback indicators. These appear in the gradebook and tell you where your work stands and what comes next.
| Indicator | Equiv Mark | Description |
|---|---|---|
| Complete without Revision | Above 70% | The work is above expectations, and no further submission is required. |
| Complete with Revision | Above 70% | An initial indicator of Revise and Resubmit was given, and the work now meets expectations. |
| Revise and Resubmit | Below 70% | The work does not meet the expectations of the task. A resubmission that acts on the feedback given is required to complete the task |
| Not Complete | Fail | If no action is taken, expectations cannot be matched, and as a last resort, a Not Complete will be awarded. A Not Complete on any Hurdle Task would result in a “Course Not Satisfied (CRN)” for the course. |
In LAWS4001, you will receive an indicator mapped to grades using the conventional scheme (HD/D/CR/P/N). You won’t see a ‘mark’ on individual items - only the final grade calculated through the official marks release process.
For LAWS students: Tasks will use quality descriptors that map to a number grade. These descriptions align with the ANU Policy on Student Assessment (Coursework). Rather than seeing a raw mark, you’ll see a description of your work’s quality, which then maps to a final numerical grade.
| Quality Descriptor | Minimum % | Likely % | Maximum % |
|---|---|---|---|
| Unacceptable | 0 | 20 | 45 |
| Unsatisfactory | 45 | 50 | 55 |
| Satisfactory | 50 | 55 | 60 |
| Satisfactory-Good | 55 | 60 | 65 |
| Good | 60 | 65 | 70 |
| Good-Superior | 65 | 70 | 75 |
| Superior | 70 | 75 | 80 |
| Superior-Exceptional | 75 | 80 | 85 |
| Exceptional | 80 | 85 | 90 |
| Exemplary | 90+ | - | - |
Here’s how this works in practice. Below is an example of how a student’s work across different tasks is assessed and converted to a final mark:
| Task/Stage | Max Wt | Quality Descriptor | Min % | Likely % | Max % |
|---|---|---|---|---|---|
| SoCIETIE Project Plan | 10% | Exceptional | 8 | 8.5 | 9 |
| KNoT completion evidence | 20% | Superior | 14 | 15 | 16 |
| Shareable Artefact Presentation | 10% | Superior-Exceptional | 7.5 | 8 | 8.5 |
| Shareable Artefact | 60% | Superior | 42 | 45 | 48 |
| Final Mark | 100% | - | 71.5 | 76.5 | 81.5 |
In this example, the student’s final mark would be around 77 (calculated from 76.5 and rounded). The final number depends on the overall evidence the teaching team sees.
Note: If you don’t submit an optional task, or if it scores lower than your main Portfolio, those marks aren’t lost - they can be redirected toward your Shareable Artefact. This is about flexibility, not penalty.
Rubric and Descriptors
Instead of a traditional rubric with point breakdowns, we describe quality by looking at what you’re actually doing with ideas and how you’re thinking about problems.
These descriptions are based on educational frameworks like SOLO Taxonomy and Bloom’s Taxonomy. They’re not rigid boxes - your work will probably show characteristics from multiple levels at once. Use these as a guide to understand what we’re looking for at each quality level.
Quality of Ideas and/or Connections
How you’re thinking about and working with concepts and information.
Used to guide grading in:
- KNoT Completion Evidence
- SoCIETIE Project Shareable Artefact
Satisfactory (Major Revisions)
Explain ideas or concepts, recall facts, concepts or answers
Associated verbs: Cite, Convert, Define, Demonstrate, Extend, Find, Identify, Interpret, Label, List, Locate, Name, Predict, Quote, Recall, Reproduce
Good (Minor Revisions)
Examine and break down information, use existing knowledge to solve new problem
Associated verbs: Analyse, Apply, Calculate, Categorise, Change, Choose, Classify, Complete, Deduce, Differentiate, Distinguish, Execute, Investigate, Operate, Practice, Relate, Select, Separate, Solve, Use
Superior (Meets expectations)
Generate new ideas, assemble novel ideas from multiple areas
Associated verbs: Assemble, Assess, Construct, Create, Design, Develop, Estimate, Generate, Invent, Measure, Plan, Predict, Produce, Synthesise, Test
Exceptional (Above expectations)
Integration of activities across levels
Quality of Narrative and/or Reflection
How clearly you explain your thinking and what you’ve learned.
Used to guide grading in:
- KNoT Completion Evidence
- SoCIETIE Project Shareable Artefact
Satisfactory (Major Revisions)
Demonstrate logical argument, clear explanations
Associated verbs: Describe, Discuss, Explain, Outline, Paraphrase, Review, Summarise
Good (Minor Revisions)
Apply knowledge in new situation, translate ideas from one domain to another
Associated verbs: Articulate, Compare, Conclude, Contrast, Correlate, Illustrate, Interpret, Show, Teach
Superior (Meets expectations)
Defend opinions and decisions, justify action through judgements about information
Associated verbs: Argue, Compose, Criticise, Debate, Defend, Decide, Evaluate, Formulate, Judge, Justify, Propose, Recommend
Exceptional (Above expectations)
Integration of activities across levels
Quality of Facilitation and/or Collaborative Team Work
How you work with others and help them do their best work.
Used to guide grading in:
- Doing KNoTs, such as projects (if relevant)
Satisfactory (Major Revisions)
Work in a functional way, convey meaning without conflict
Associated verbs: listen, notice, tolerate, comply, enjoy, follow, build, perform, execute, implement, copy, follow, replicate, repeat
Good (Minor Revisions)
Build on strengths of individuals for the benefit of the whole
Associated verbs: express, conduct, show, demonstrate, complete, perfect, control
Superior (Meets expectations)
Extend strengths, enable others to produce their best work
Associated verbs: amplify, choose, consider, prefer, discriminate, depict, exemplify, construct, solve, integrate, adapt, enable, influence
Exceptional (Above expectations)
Integration of activities across levels
Submission Windows
You’ll see submission windows associated with all tasks. We encourage you to submit within the window. This is intentional:
- Submit before the window = early, usually rewarded with feedback
- Submit during the window = on time, usually accepted as complete/incomplete
- Submit after the window = late
In most cases, submissions stay open after the due date, so don’t stress if you’re a day or two behind, but do let the convenor know what’s going on - communication is more important than the deadline.
If you’re running late: Reach out to the convenor before the due date. Most of the time, an extension is fine, especially if it doesn’t affect other students. Just ask - it’s better than submitting late without talking to us first.
Working with AI
The SoCIETIE Initiative does not ban AI. We ask you to be honest about it.
One of our inaugural cohort put it better than we could. Jake described AI as his collaborator, critical friend and business partner - something that gave him Superhuman capability. We think that this perspective matters most when working with Community organisations who already run on slim resources. Working with these tools might be a useful thing to be able to do, and we’d rather you learned it here.
So the question we’re interested in is not did you use AI. It’s what did you bring to it, and what did you learn.
What we ask
Declare it, every time. Most submissions will ask how you developed the work. Tick everything that applies:
- Entirely my own work - no AI, no collaborators. Ticking this clears the others, because it’s a claim that contradicts them.
- AI as assistant - you used it around the edges: finding sources, summarising reading, checking your logic, tidying your writing. The thinking and the words are yours.
- AI as collaborator - you worked with it closely, and parts of what you’re submitting were generated with AI that you then directed, checked and shaped.
- Peers as collaborators - you worked closely with others to develop this.
- Something else - with a box to tell us what.
You can tick more than one, because most honest answers are more than one. Working with AI and thinking something through with two classmates is an ordinary way to make good work, and we’d rather see both than make you choose which to mention.
None of these is the right answer. We are not scoring the boxes you tick. We’re asking because we don’t yet fully understand what good practice looks like in this space, and the honest answers of a cohort working on real community problems are the best evidence any of us will get.
Peers are on that list for the same reason AI is. This course is built on working with other people; it would be strange to treat that as something to leave out.
Be able to stand behind it. Whatever the box says, you are accountable for what you submit: every claim, every source, every number. “The AI said so” is not a defence, and it won’t be one after you graduate either. AI tools fabricate confidently - invented citations, plausible-looking statistics, quotes nobody said. Check anything you’d be embarrassed to be wrong about.
Protect other people’s information. This one is specific to us. You are working with real community partners, and much of what they tell you is theirs, not yours. Don’t paste partner data, personal details, unpublished material, or anything shared in confidence into a public AI tool. If you’re unsure whether something is yours to share, it isn’t - ask us.
Why we can be progressive about this
Most of what SoCIETIE assesses is your reflection on what you did - the KNoT you turned up to, the project you ran, the partner you worked with, what changed in your thinking. This is the learning. That evidence is yours and nobody else’s, so there is very little here worth outsourcing. Even if you outsource the task, you still should be learning - how to write better prompts, how to be more efficient, how to get better answers - this is what we want to support.
The wider guidance
The ANU position is set out in the ANU Institutional AI Principles, approved by Academic Board in 2023. Two of the six sit directly behind this section, and in tension: the commitment to “excellence and integrity in teaching, learning, assessment”, and the commitment to “produce graduates with the knowledge and skills to operate effectively and ethically in an AI-informed world”.
This section covers SoCIETIE assessment only. Your other courses set their own parameters to meet their learning outcomes.
Citations and Referencing
The key guide for refencing and citation in this course is a decision for you. Either reference the way your discipline references or reference the way that works in your project/artefact. We don’t mandate a style. You are here from law, economics, engineering, psychology, policy, science, the arts and everything in between - and the referencing system you already use is the one you should keep using.
A citation style is a discipline’s argument about what matters in a source, compressed into punctuation. Lawyers pinpoint the paragraph because the paragraph is the authority. Engineers number their sources because the claim matters more than the person making it. Psychologists lead with the year because a 1998 finding and a 2024 finding are different kinds of evidence. Noticing those differences is noticing something real about how each field decides what counts as true - which is most of what transdisciplinary work is.
| If you come from… | ..You probably use… | ..Which tells you… |
|---|---|---|
| Law | AGLC - footnotes with pinpoint references | The exact paragraph is the authority; a reader must be able to land on it |
| Economics | Author-date (Harvard, Chicago author-date) | Working papers and data releases matter; recency and revision stay visible |
| Public policy | Harvard or APA, heavy on grey literature | Reports, submissions and government documents carry the argument, not just journals |
| Engineering | IEEE - bracketed numbers | The claim stands on its own; authorship is secondary to the result |
| Mathematics | AMS - numbered or [Abb-Year] | Results are permanent and attributable; who proved it first is the point |
| Psychology | APA 7 - author-date | Method and date are load-bearing; the reader is assessing replicability |
Pick one and be consistent within a piece of work. If you’re genuinely between disciplines, choose the one your reader is most likely to come from - which for a SoCIETIE artefact is often your community partner rather than an academic.
If your artefact isn’t a written piece, you still credit. A podcast has show notes or a spoken acknowledgement; a video has an end card; a zine has a colophon; an exhibition has a wall label. The form changes, the obligation doesn’t. Whatever you made, someone should be able to find out where the ideas came from.
Indigenous knowledges
Indigenous knowledges are welcome in your work, and they are not always citable the way a journal article is. Knowledge may be held by a custodian rather than authored by a writer, shared orally rather than published, belong to a community rather than an individual, tie to a specific Country, or carry conditions on how and whether it may be shared at all.
Use the Indigenous Knowledges Attribution Toolkit (IKAT), developed by the Indigenous Archives Collective with CAVAL. It has a decision tree for working out what you are dealing with and how to attribute it, plus worked examples - including how to name a knowledge holder, their nation or community, and the context in which knowledge was shared. The citation guide is a PDF you can keep open while you write.
Before you cite, ask whether you may
Attribution and permission are different things. Some knowledge is not yours to reproduce even with a perfect citation, and a correct reference does not turn something shared in confidence into something published. If you learned something from a person - an Elder, a knowledge holder, a partner, anyone - ask them how they want to be credited, and whether they want it included at all. Their answer settles it. Talk to us if you’re unsure.
Citing AI
Declaring AI on the submission form is not the same as citing it in the work. The declaration tells us how you worked. A citation tells your reader which words or ideas came from a machine. If you quote AI output, reproduce it, or lean on it for a specific claim, cite it where it appears.
The major styles have landed in roughly the same place, with one useful exception:
- APA 7 treats the developer as the author - Anthropic, OpenAI, Google - names the tool and version, and asks for a shareable link to the specific conversation where the tool offers one (APA’s own guidance).
- Harvard does much the same: creator as author, with tool name, version and model type in square brackets, and a shareable URL where possible.
- IEEE asks you to disclose AI-generated content - text, figures, images or code - in an acknowledgements section.
Two practical notes. AI output is not retrievable: your reader cannot follow the reference and see what you saw, which is precisely why the styles handle it awkwardly. Keep your transcripts - if a claim is ever questioned, that record is what you have. And if the AI gave you a source, go and read the source and cite that, not the AI. Fabricated references are the most common way this goes wrong, and citing a paper you haven’t opened is a problem whether or not a machine suggested it.
Version Control
Author: Chris.Browne@anu.edu.au Last updated: 06-Aug-2026